TrueForge open-source agent harness cuts costs by 30%-75% in testing
omarsar0 · x · 2026-08-20
TrueFoundry released TrueForge, an open-source (MIT) agent runtime harness designed to be vendor-neutral. It manages tool-calling loops, context, sub-agents, and sandboxed code execution across OpenAI, Anthropic, Google, and open-weight models like Kimi, GLM, and DeepSeek. In a 14-task enterprise benchmark, TrueForge driving Opus 4.8 matched Claude Managed Agents' accuracy at 30% lower cost (3.8M vs 10M tokens). Routing to GLM-5.2 maintained accuracy while cutting costs by 75% ($3 vs $12).
More from coding & agent
- Gemini 4 Argon goes live in Antigravity with its own weekly quota for Pro subscribers — rseroter · 2026-10-16
- LangChain revamps human-in-the-loop middleware with typed schemas for tool interrupts — Hacubu · 2026-10-16
- Robot fully controlled by frontier LLMs assembles gadgets autonomously for 1.5 hours error-free — ihorbeaver · 2026-10-16
- Why the future of agents is basically Node and JavaScript, in one argument — verrsane · 2026-10-16
- Vibe Coding a Data Pipeline Is Easy; Running It at Billions of Rows Is Not — Zachly · 2026-10-16
- How Claude Code Review works at GA: admin setup, auto reviews on every new PR — ClaudeDevs · 2026-10-16